Multimodal Face Authentication Using Adaptive Radar and Camera Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional biometric authentication systems, particularly those relying on vision, face challenges in poor lighting conditions, leading to inaccurate user verification and potential unauthorized access.
Innovation Solution
A multimodal authentication system that combines millimeter wave radar and camera data to generate facial signature and image data, with a processor assigning weights based on illumination and sensor conditions to ensure reliable verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If vision-based biometric authentication is used, then authentication speed is fast, but authentication accuracy deteriorates under poor lighting conditions
Solution Approach 1:
The patent combines radar-based facial recognition with vision-based camera-based recognition into a unified authentication system. The radar module captures facial depth information and 3D structural data, while the camera captures 2D facial images. Both modalities are processed and fused to generate a comprehensive authentication result, allowing the system to maintain high accuracy under varying lighting conditions by relying on radar data when vision is compromised.
2Reliability
If multiple sensors are added to improve authentication reliability, then authentication accuracy improves, but device complexity increases
Solution Approach 1:
The radar module serves multiple functions: it captures facial depth information, provides 3D structural data for authentication, and can operate independently or in conjunction with the camera system. This multi-functionality allows the system to improve authentication reliability through radar capabilities without proportionally increasing overall system complexity, as the same radar hardware supports both standalone and fused authentication modes.
3Reliability
If radar data is used for authentication, then authentication accuracy improves under poor lighting, but processing time increases
Solution Approach 1:
The system implements adaptive authentication strategies where it can process only the necessary modalities based on environmental conditions. When lighting is adequate, the system may rely primarily on faster camera-based authentication. When lighting is poor, it activates radar-based authentication which provides superior accuracy but requires more processing time. This selective approach optimizes the balance between speed and accuracy by applying partial processing based on contextual needs.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances authentication accuracy by leveraging radar and camera data synergistically, improving reliability under varying lighting conditions and sensor constraints, thereby preventing unauthorized access.
Implementation Method 1
a radar source configured to generate a first input
Data Source
AI summary
An electronic device, a method and computer readable medium for multimodal authentication are disclosed. The method includes receiving a request for authentication, facial signature data generated based on an input from a radar source of the electronic device, and facial image data generated based on an input from a camera of the electronic device. The method also includes identifying an illumination condition and a sensor condition associated with the electronic device. The method additionally includes assigning a weight associated with the camera and the radar source based on the illumination condition and the sensor condition. The method further includes granting the request when at least one of the facial signature data and the facial image data are within a threshold of similarity with a preregistered facial data associated with the weight.